Zhang, Yao, authorPaustian, Keith, advisorArabi, Mazdak, committee memberParton, William, committee memberSchipanski, Meagan, committee member2016-08-182016-08-182016http://hdl.handle.net/10217/176724To provide better understanding of natural processes and predictions for decision support, dynamic models have been used to assess impact of climate, soils and management on crop production, water use, and other responses from field to regional scales. It is important to continue to improve the prediction accuracy and increase the reliability. In this work, we first improved the DayCent ecosystem model by developing a new empirical method for simulating green leaf area index (GLAI) of annual crops. Its performance has been validated using experimental observations from different experimental field locations as well as more aggregate NASS yield data spanning the country. Additionally, sensitivity and uncertainty of important parts of the crop growth model have been quantified. Our results showed the new model provided reliable predictions on crop GLAI, biomass, grain yield, evapotranspiration (ET), and soil water content (SWC) at field scale at various locations. At national scale, the predictions of grain yields were generally accurate with the model capable of representing the geographically-distributed differences in crop yields due to climate, soil, and management. The results indicated that the model is capable of providing insightful predictions for use in management and policy decision making. Although there are challenges to be addressed, our results indicate that the DayCent model can be a valuable tool to assess crop yield changes and other agroecosystem processes under scenarios of climate change in the future.born digitaldoctoral dissertationsengCopyright and other restrictions may apply. User is responsible for compliance with all applicable laws. For information about copyright law, please see https://libguides.colostate.edu/copyright.Text